memory-checkpoint

Save and retrieve session state via semantic search and markdown checkpoints.

29|2|Updated Jan 11, 2026

Search Tags:#versioning #context management #memory #continuity #checkpoint #session state

Ready to run

npx skills add https://github.com/jbdamask/john-claude-skills --skill memory-checkpoint-jbdamask

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: memory-checkpoint
Source: https://github.com/jbdamask/john-claude-skills/tree/main/skills/memory-checkpoint
Command: npx skills add https://github.com/jbdamask/john-claude-skills --skill memory-checkpoint-jbdamask

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires fastembed, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill allows you to save the current state of your development session and restore it later, ensuring continuity even when Claude's context window is compressed or reset.

Core Features & Use Cases

Save Session State: Create a snapshot of your current work, including progress, decisions, and key files.
Search Past States: Find previous checkpoints based on semantic similarity to a search query.
Use Case: You're working on a complex feature and need to switch tasks. Save your current progress with /memory-checkpoint. Later, you can recall it or search for a specific past state like /memory-checkpoint authentication bug to find relevant past work.

Quick Start

Use /memory-checkpoint to save your current session state.

Frequently Asked Questions about memory-checkpoint

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I save session state for continuity across context compressions?▼

To save session state for continuity across context compressions, you can create a snapshot of your current work using a checkpoint command, which stores your progress, decisions, and key files locally for later restoration.

How does semantic search work for finding previous development checkpoints?▼

Semantic search for finding previous development checkpoints works by indexing saved markdown state representations using the fastembed library, allowing you to retrieve past project states based on semantic similarity to a search query.

What is the best way to manage project state when switching tasks?▼

The best way to manage project state when switching tasks is to save your current progress as a structured markdown checkpoint, which can later be searched and restored to resume your work exactly where you left off.

Does fastembed support semantic similarity search for local session state?▼

Yes, fastembed supports semantic similarity search for local session state by indexing structured markdown checkpoints, enabling the retrieval of relevant past development work based on natural language queries.

Can I use markdown parsing to represent and restore development session state?▼

Yes, you can use markdown parsing to represent and restore development session state, as the system leverages markdown parsing to structure and store project state snapshots for later retrieval and continuity.

When should I use a checkpoint to manage context window resets?▼

You should use a checkpoint to manage context window resets when you are working on a complex feature and need to switch tasks, ensuring your current progress and decisions are saved locally before the context is compressed or cleared.